Diffusion-driven distillation and contrastive learning for class-incremental semantic segmentation of laparoscopic images.

ABSTRACT:

Purpose

Understanding anatomical structures in laparoscopic images is crucial for various types of laparoscopic surgery. However, creating specialized datasets for each type is both inefficient and challenging. This highlights the clinical significance of exploring class-incremental semantic segmentation (CISS) for laparoscopic images. Although CISS has been widely studied in diverse image datasets, in clinical settings, incremental data typically consists of new patient images rather than reusing previous images, necessitating a novel algorithm.

Methods

We introduce a distillation approach driven by a diffusion model for CISS of laparoscopic images. Specifically, an unconditional diffusion model is trained to generate synthetic laparoscopic images, which are then incorporat

SUBMITTER: Zhao X 

PROVIDER: S-EPMC12226607 | biostudies-literature | 2025 Jun

REPOSITORIES: biostudies-literature